[cfe-commits] r141053 - in /cfe/trunk: include/clang/Basic/DiagnosticDriverKinds.td lib/Driver/Tools.cpp

Bob Wilson bob.wilson at apple.com
Tue Oct 4 10:24:35 PDT 2011


On Oct 4, 2011, at 9:52 AM, Ted Kremenek wrote:

> On Oct 4, 2011, at 9:03 AM, Bob Wilson wrote:
> 
>> I'm not sure this is a good idea.  I understand the concern about having -W flags for all warnings, but we certainly don't anyone to start using -Wno-unsupported-debug-with-integrated-as.
> 
> Can you clarify your concern?  If people know what they are doing, they should be able to silence the warning (or promote them to an error).

Yes, I guess it's a tradeoff.  I'm just envisioning the scenario where someone is annoyed by this warning, changes a makefile to disable it, and then gets a build failure later when we remove the warning option.  This warning is just a temporary hack to alert people to an unimplemented feature.

> 
>>  As soon as the integrated assembler can support debug info, we're going to rip out this warning, and we don't want to be stuck keeping the -W flags around for backward compatibility.
> 
> I don't see that as an issue.  We never promise that warnings will stay around forever.  Once a warning is deemed not to be useful anymore we should feel free to remove it.
> 
>>  Is there some existing warning group that we could use for this?
> 
> No, and I don't think that would be the right approach anyway.  That seems counter to your concern about putting this warning under a flag in the first place.
> 
> If you really don't want driver warnings to be under a -W flag, then we should consider adding another diagnostic category to segregate them from other warnings.  Once we get all warnings under a -W flag, we'd like to change tablegen to reject warnings that are not in a diagnostic group.

Right.  I was wondering if we should have something like GCC's "notes" (maybe we already do?)  But, then you still have the problem that someone may want to turn it off, so I don't think that's the right thing, either.

It's fine to leave this as it is.  I think it's pretty unlikely that anyone is going to add -Wno-unsupported-debug-with-integrated-as to makefiles.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20111004/2e28ea71/attachment.html>


More information about the cfe-commits mailing list